Journal of Reproductive Immunology, 22 (1992) 269-279

269

Elsevier ScientificPublishers Ireland Ltd. JRI 00786

Distribution of lymphocyte subsets in rat milk from normal and Trichinella spiralis-infected rats H y u n g R . N a a, J o h n C. H i s e r o d t b a n d L e o n a r d

L. Seelig, J r a

aDepartment of Cellular Biology and Anatomy, Louisiana State University Medical Center, Shreveport, LA 71130 and bDepartment of Pathology, University of Pittsburgh, PA 15213 (USA)

(Accepted for publication 6 May 1992)

Summary We have shown that T. spiralis-specific T lymphocytes can mediate maternal-toneonatal immunity during lactation. This study addresses the change of lymphocyte populations in rat milk during normal and disease conditions. Two color flow cytometric analysis was performed for milk lymphocytes. T cells (OX19 +) made up 45% of rat milk lymphocyte population. T helper cells (Th) composed 35% of total T cells while T cytotoxic/suppressor (Tcs) cells constituted 34%, giving a Th/Tcs ratio of 1,03. The corresponding ratio Th/Tcs in peripheral blood was 2.8. Approximately 21% of OX8 + cells in rat milk were OX19- natural killer (NK) cells. When using the monoclonal antibody 3.2.3 (NKR-P1), 43% of lymphocytes in control rat milk and 14% of blood lymphocytes were N K cells. This indicates a selective passage of these cells into milk. In T. spiralis-immunized rats, the percentage of total T cells was slightly decreased; however, Th and Tcs cells were consistent as compared to control milk. The percentage of N K cells (OX8÷OX19 -) in milk from T. spiralis-immunized rats was significantly higher than that from control milk (65% vs. 21%, respectively, P < 0.01). This result was confirmed using the monoclonal antibody 3.2.3 which showed that milk from immunized rats contained 63% N K cells compared to 43% in normal milk (P < 0.01). This study suggests that N K cells are selectively passaged into rat milk and T. spiralis infection induces an increase of N K cells in milk.

Correspondenceto."Dr. Leonard L. Seelig,Jr., LSU Medical Center at Shreveport,Departmentof Cellular Biology & Anatomy, P.O. Box 33932, Shreveport, LA 71133-3932, USA.

0165-0378/92/$05.00 © 1992 Elsevier ScientificPublishers Ireland Ltd. Printed and Published in Ireland

270

Key words: milk lymphocytes; NK cells; Trichinella spiralis; flow cytometric

analysis

Introduction

The presence of leukocytic cells in colostrum and milk has been confirmed in humans (Diaz-Jouanen and Williams, 1974) and in animals (Lee et al., 1980) including rats (Head and Beer, 1978). Lymphocytes constitute approximately 10% of human and rat milk cells (Crago et al., 1979; Head and Beer, 1978). The distribution of various T cell subpopulations such as T helper (37%), T cytotoxic/suppressor (34%) and TCR-3~6 cells (11%) in human colostrum has been reported (Richie et al., 1982; Berttoto et al., 1990). Natural killer cell-enriched populations of human milk cells has been shown to have cytotoxic activity, although not to the extent seen in peripheral blood (Moro et al., 1985; Nair et al., 1985). The higher proliferative response of milk lymphocytes to enteric antigens, compared to those of blood, have suggested that these cells are a selected population of immunocompetent cells derived by induction at a distant anatomical site, such as gut-associated lymphoid tissue, which then circulate to the mammary gland during pregnancy and lactation (Parmely et al., 1976; Goldman et al., 1982; Oksenberg et al., 1985). The passage of leukocytes through the alveolar wall to enter milk has been shown by Seelig and Beer (1978). There is now considerable evidence to support the theory that milk cells can transmit immunity to the neonate during lactation. Studies by Ogra et al. (1977) have shown that infants breast-fed by tuberculin-positive mothers develop positive peripheral blood proliferative responses to tuberculin, and Archambault et al. (1988) has demonstrated that orally fed colostral lymphocytes from immune animals can transfer protection against rotavirus infection to newborn calves. The enteric parasite, T. spiralis has been shown to provoke both specific and non-specific immune response in the infected host (Wakelin and Denham, 1983) and the transfer of T. spiralis-specific immunity, with milk whey, serum IgG (Appleton and McGregor, 1987) and sensitized T helper cells (Kumar et al., 1990) from mothers to the suckling neonates via lactation has been reported. Immunohistochemical staining of leukocytes (Kumar et al., 1991) has shown that in T. spiralis-infected animals, there is an increased number of cell surface marker OX8 positive cells (T cytotoxic/suppressor cells and natural killer cells) and a decreased number of T cells in the epithelium of rat mammary gland, indicating a selective passage of certain cell types into milk upon immune stimulation.

271

Although some studies have tested the humoral immune components in rat milk during antigenic stimulation (Steele and Leslie, 1987), no studies have analyzed the cellular components of milk in control or antigen-stimulated rats. In this study, using monoclonal antibodies specific for T cells, their subsets, N K cells and flow cytometric analysis, we have determined the distribution of the subpopulations of lymphocytes and the effect of immune stimulation by T. spiralis infection on cellular populations in rat milk. Materials and Methods

Animals Rats of the Fischer (FI) strain, housed in temperature- and lightcontrolled rooms and provided laboratory chow and water ad libitum, were used in these experiments. Three-week-old virgin females were infected with the parasite T. spiralis, mated 3 weeks later and milked on day 4 - 6 postpartum. Age-matched uninfected animals served as controls in these experiments.

Milk cell preparation Neonates were separated from their mothers 3 h prior to milking. Rats were anesthetized with 3.6% chloral hydrate (0.1 cc/100 g body wt) and given an injection of 2 units Pitocin (Parke-Davis), and milk was collected by a procedure established in our laboratory using a customized milking machine that enables the simultaneous collection of milk by vacuum from all 12 rat nipples (approximately 4 - 6 ml/rat). Pooled milk was three-fold diluted with Ca 2+ and Mg2+-free Hanks balanced salt solution (Sigma), centrifuged at 400 x g for 20 min at 4°C, washed twice and resuspended at approximately 1 × 10 6 particles/ml. At this stage, the milk sample contains a small proportion of cells and a large number of milk fat globules.

Isolation of T. spiralis larva and infection of rats Muscle from T. spiralis-infected mice was placed in a Waring blender in 1 1 of a mixture of 1.0% pepsin and 10% HC1, minced for 20 s and incubated at 37°C for 1 h. The suspension was filtered through gauze into Imhoff funnels and the Ll-stage larva were allowed to settle for 10 min. The pepsin HC1 solution was removed by suction, the worms were resuspended in 1 1 of 0.85% saline and allowed to resettle for 10 min. Concentrated larva were removed from the funnels and suspended in saline at a concentration of 1000 larva/ml. Viability was assessed as coiled worms for infection. Rats were lightly anesthetized with metaphane and infected with 1000 viable larva by gastric intubation. Success of infection was monitored in each animal at autopsy by

272

counting T. spiralis worms isolated from the female's intestine according to the procedure of Stewart et al. (1985).

Antibodies Fluorescein isothiocyanate (FITC)-conjugated mouse anti-rat monoclonal antibody OX1 (CD45) for total leukocytes, R-phycoerythrin (RPE)conjugated OX19 (CD5) for total T cells, W3/25 (CD4) for T helper cells (Th) and OX8 (CD8) for T cytotoxic/suppressor cells (Tcs) were obtained from Bioproducts for Science, IN. FITC-conjugated F(ab')2 fragments of monoclonal antibody (mAb) 3.2.3 (NKR-P1) for natural killer cells was kindly provided by Dr. Hiserodt, Univ. of Pittsburg. FITC-conjugated goat anti-mouse IgG was obtained from Jackson Immunoresearch Lab., PA.

Flow cytometric analysis An Epics 753 flow cytometer/cell sorter (Coulter Cytometry, Hialeah, FL), equipped with a 5 W argon laser, a 4 W argon laser and a dye laser was used for the simultaneous accumulation of two-color immunofluorescence in addition to forward angle and right angle laser light scatter signals. FITC and PE signals were separated using filters and the overlap in the emission spectra of FITC and PE was compensated electronically. Lymphocytes were gated on the basis of their size (forward light scatter) and internal granularity (right angle light scatter). Normally, 5 x 10 3 to 1 × 104 cells were accumulated for analysis and the percentage of positive cells was subtracted by the percentage of negative controls above the gating channel.

Staining procedure Two-color analysis was done for the enumeration of T cells and their subsets to remove the effect of non-specific staining of milk fat globules. All antibodies were diluted at appropriate concentration with phosphate buffered saline (PBS) containing 2% fetal bovine serum and 0.1% sodium azide. For determination of relative number of T cells, cells (1 × 106) were stained with 200 #1 of FITC-OX1 (1:50) and RPE-OX19 (1:10). For T cell subsets, cells were stained with the following antibodies, sequentially; (1) W3/25 (1:100) for Th cells or OX8 (1:100) for Tcs cells; (2) FITC-goat anti-mouse IgG (1:400); (3) 10% normal mouse serum; and (4) RPE-OX19. For natural killer cells, OX1, RPE-goat anti-mouse IgG, normal mouse serum and FITC-3.2.3 (1:10) were used sequentially. All staining procedures were performed on ice for 30 min and cells were washed twice with buffer solution between each step. The stained cells were fixed with 1% paraformadehyde in PBS, stored at 4°C in the dark and analyzed within 1 week. Unstained cells, cells stained with FITC- or RPE-goat anti-mouse IgG, or with an appropriate isotype antibody were used as negative controls.

273

Data and statistical analysis The ratio of total T cells to total lymphocytes, T subsets to total T cells and natural killer cells to total OX8 ÷ cells or to total lymphocytes were calculated as a percentage. Differences between groups were analyzed using the Wilcoxon rank sum-test or the Wilcoxon signed rank test (Statistix II, NH Analytical Software, Roseville, MN). Results

Rat milk contains numerous fat globules which are lipid-laden, anucleate cell fragments derived from mammary secretory epithelial cells. After ceils were treated with a density gradient procedure, a large number of fat globules were still found in the cell suspension and contributed to considerable non-specific background staining. Therefore, two color flow cytometric analysis was done after making a lymphocyte gating bit map on the basis of forward and right angle scatter. Since there was no cell surface marker for total rat lymphocytes, cells were stained with antibody OX1 (total leukocytes) and analyzed only inside the delineated lymphocyte bit map. Using this procedure, the OX1 positive cells in a bit map represented the lymphocyte population. After rat milk cells were stained with FITC-conjugated OX1 and PEconjugated OX19 (T cells) antibody, two color flow cytometric analysis showed that T cells (OX19 ÷) made up approximately 43% of this gated rat milk lymphocyte population (Table 1). The Th phenotype (W3/25 +) composed approximately 35% of total T cells while the Tcs (OX8 ÷) phenotype constituted 34%, given a Th/Tcs ratio of 1.03. The Th/Tcs ratio of rat milk was lower than that of corresponding control rat blood (2.5). A similar decreased ratio of Th to Tcs cells in human milk compared to that of blood has also been reported (Richie et al., 1982). Rat natural killer (NK) cells have

TABLE 1

245 Percentage of lymphocyte subsets in rat milk (n = number of pooled milk samples). T cells (OX19/OX1)

T helper

(W3/25/OXI9)

T cytotoxic (OXS/OXI9)

N K cells (OX8 +OXI9-/OX8)

Non-infected control (n = 5)

43 ± 7 a

35 ± 3

34 ± 5

21 + 5

T. spiralis

29 ± 4

42 -~ 3

39 ± 8

65 ± 8*

infected (n = 5) aMean ± standard error o f mean. *Significantly different from non-infected control group at P < 0.01.

274

B

A o-=

+

C3-

",;